The molecular mechanisms underlying cancer have been mainly studied one or a few "cancer genes" at-a-time. However, it is thought that combinations of mutated or aberrantly expressed tumor suppressor- and oncogenes may be responsible for advancing cells through most steps of tumorigenesis. Many cancer causing mutations are disrupting interactions and these alterations are often directly related to the mechanism of pathogenesis. Thus, altered protein-protein interactions may directly point to a mechanism for cancerogenesis. More importantly, since it is becoming increasingly clear that genes and their products interact in complex biological networks with local and global properties, it is possible that perturbations of these networks contribute to cancer formation. We propose that a further understanding of the mechanisms involved in cancer, and the development of new therapeutic strategies, can be gained by i) studying genes and their products in the context of the molecular networks in which they function, and ii) investigating how such networks are altered in tumor cells compared to their unaffected counterparts. In addition to the information available from several drafts of the human genome sequence, genome- wide experimental strategies have been developed that will help us understand the effects of cancer mutations in the context of molecular networks: i) protein-protein and DNA-protein interaction networks or "interactome" networks are being mapped at an increasing pace, producing datasets with ever increasing quality and decreasing costs, and ii) large numbers of cancer-associated mutations are being discovered in the context of the human cancer genome project. Here we propose to develop a genome-wide application for a new technology platform that we have recently initiated to systematically study the effects of cancer-associated mutations on the physical and functional interactions mediated by the products encoded by cancer genes in the context of global interactome models. Our specific aims are to apply our experimental and computational technology platforms to: i) clone large numbers of cancer-associated missense or single amino acid change (SAC) alleles, ii) identify and characterize the interaction properties of large numbers of SAC alleles, iii) analyze the effects of SAC alleles on the local and global properties of interactome networks. [unreadable] [unreadable] [unreadable]